Rethink the Future of AutoML
A brief discussion about AutoML at present and in the future.
Dec 10, 2024
Architecture Modeling and What It Empowers
A brief discussion about architecture modeling in neural architecture search.
Dec 10, 2023
Dynamic Ensemble of Low-Fidelity Experts: Mitigating NAS "Cold-Start"
Leverage the beneficial knowledge implicit in low-fidelity information (e.g., one/zero-shot estimation) to mitigate the "cold-start" problem of predictor-based NAS.
Jun 26, 2023
TA-GATES: An Encoding Scheme for Neural Network Architectures
This work proposes a new encoding scheme for neural architectures, the Training-Analogous Graph-based ArchiTecture Encoding Scheme (TA-GATES). TA-GATES encodes an NN architecture in a way that is analogous to its training.
Nov 28, 2022
Discovering Robust Convolutional Architecture at Targeted Capacity: A Multi-Shot Approach
Considering scenarios with capacity budget, we aim to discover adversarially robust architecture at targeted capacities. However, one-shot NAS methods favor topologies with larger capacities in the supernet. And the discovered topology might be suboptimal when augmented to the targeted capacity. We propose a novel multi-shot NAS method to address this issue and explicitly search for robust architectures at targeted capacities.
Dec 22, 2020